Searching the Higgs with the Neurochip TOTEM

TL;DR

This paper demonstrates that neural network classifiers, specifically the TOTEM neurochip trained with RTS, can effectively distinguish Higgs production events from background noise at the LHC for a Higgs mass around 200 GeV.

Contribution

It introduces the use of the TOTEM neurochip with RTS training for real-time Higgs event classification at the LHC, a novel application of neurocomputing in high-energy physics.

Findings

Neural networks can discriminate Higgs events from background.

The TOTEM neurochip performs effectively in on-line event classification.

Comparison of different input variable sets shows their impact on classification accuracy.

Abstract

We show that neural network classifiers can be helpful in discriminating Higgs production events from the huge background at LHC, assuming the case of a mass value $M_H \sim 200$ GeV. We use the high performance neurochip TOTEM, trained by the Reactive Tabu Search algorithm (RTS), which could be used for on-line purposes. Two different sets of input variables are compared.